Porous hydroxyapatite-based obturation materials

for dentistry
Witold Brostow
a)
Laboratory of Advanced Polymers & Optimized Materials, Department of Materials Science
and Engineering, University of North Texas, Denton, Texas 76203-5310
Miriam Estevez
Centro de Fsica Aplicada y Tecnologa Avanzada, Universidad Nacional Autnoma de Mxico,
Quertaro, Qro. 76000, Mxico
Haley E. Hagg Lobland and Ly Hoang
Laboratory of Advanced Polymers & Optimized Materials, Department of Materials Science
and Engineering, University of North Texas, Denton, Texas 76203-5310
J. Rogelio Rodriguez
Centro de Fsica Aplicada y Tecnologa Avanzada, Universidad Nacional Autnoma de Mxico,
Quertaro, Qro. 76000, Mxico; and Centro de Investigacin y Estudios Avanzados del Instituto
Politcnico Nacional, Quertaro, Apdo. Postal 1-798, Qro., 76001, Mxico
Susana Vargar
Centro de Fsica Aplicada y Tecnologa Avanzada, Universidad Nacional Autnoma de Mxico,
Quertaro, Qro. 76000, Mxico
(Received 28 August 2007; accepted 4 January 2008)
New porous biomaterials based on hydroxyapatite (HAp) were designed as obturation
materials for dental cavities. Synthetic HAp powder with a particle diameter of
150 m was agglutinated using three different polyurethane monocomponents (rigid,
semi-rigid, and flexible), enabling the matching of their properties to those of real
teeth. Alumina particles were also added in some cases. Our new hybrid materials
contain up to 60% HAp. Interconnected pores range in size from 100 to 350 m,
while the pore volume fraction varies between 25% and 60%. Most of these materials
possess the right morphology for implants and prostheses because their porous
structures can be vascularized for bone and tooth ingrowth. Some samples also contain
alumina particles to improve the abrasion resistance and to support the stresses
produced during mastication. The materials were characterized by x-ray diffraction,
scanning electron microscopy, and mechanical testing, along with abrasion, scratch,
sliding wear, friction, and staining tests.
I. INTRODUCTION
Numerous different types of biomaterials have been
designed for human implants or prostheses, specifically
for bone repair or bone ingrowth (e.g., traumatism and
bone disease). However considerably less effort has been
addressed to implants capable of tooth ingrowth.
13
For
tooth and bone implants, a primary requirement is that
the material be bioaccepted, because vascularization re-
quires the material to support cellular activity without
eliciting an inappropriate host response on recognition of
the foreign molecules (i.e., molecular recognition).
4
Sec-
ond, the morphology must be suitable to allow vascular-
ization and attachment to the existing bone or tooth sub-
strates.
5,6
Morphological specifications for bone-implant mate-
rials require a pore size in the range of 100400 m, with
the pores being interconnected and comprising a volume
fraction between 50% and 70%.
7
For tooth implants, on
the other hand, the mean optimal pore size is 2.90
0.22 m (standard deviation), which is considerably
smaller.
8
A precise densitometry study
9
showed that
enamel has a density distribution that is narrower with
respect to that of dentin: namely, the density of enamel is
reportedly between 2.49 and 3.00 g/mL (mean density
2.94 0.03 g/mL), while the density of dentin is between
2.06 and 2.24 g/mL (mean density 2.14 0.01 g/mL).
Using these values, it is possible to demonstrate that the
pore volume fraction of dentin is around 27%, a value in
good agreement with that of 22%, which was obtained
a)
Address all correspondence to this author.
e-mail: brostow@unt.edu
DOI: 10.1557/JMR.2008.0191
J. Mater. Res., Vol. 23, No. 6, Jun 2008 2008 Materials Research Society 1587
using results that have been reported elsewhere.
8
One
observes then that pore size and pore volume fraction are
smaller for dentin than for bone.
712
While porous materials have many important applica-
tions in, for example, acoustic and thermal insulation,
transportation, filtration, purification, biomaterials,
building constructions,
13
a new generation of porous
biomaterials has recently emerged enabling one to bet-
ter reproduce the structure of natural bone. Several
groups
14,15
have been successful in controlling the size,
volume, and interconnectivity of pores in their materials.
Nevertheless, there have been only a few reports
1618
on
the use of porous materials for dental fillings: a possible
reason for this is that porous materials tend to have poor
mechanical properties, while the mastication process pro-
duces high compression and shear stresses that must be
supported by the obturation material. Consequently, non-
porous materials, usually hard polymer resins, are more
commonly used to support these stresses. However, we
claim in this article that, by the selection of an appropri-
ate agglutinating polymer (for a ceramic filler), it is pos-
sible to create porous materials with suitable morphology
and mechanical strength.
Beyond achieving the correct morphology, a success-
ful dental obturation material must be chemically com-
patible with and adhere to the substrate.
1923
Many ob-
turation materials are designed essentially by controlling
only mechanical properties because adding a ceramic, or
generally a filler, is known to improve the mechanical
properties of polymers.
24
Here, we take into account
morphology, chemical structure, mechanical behavior,
and surface properties, considering the combined effect
of all of these on tooth ingrowth,
2527
as well as the role
of viscoelasticity for implant compliance and perfor-
mance.
28,29
Because teeth constitute an organicinorganic hybrid,
a reliable hybrid
30
for dental applications
31
should con-
tain an agglutinating polymer that possesses: (i) high
shear strength (around 70 MPa) resistance interfacial
stresses during mastication; (ii) appropriate tensile/
compressive strength and toughness, because very rigid
and tough materials may lead to premature wear of the
real teeth from contact during chewing
32
; (iii) high
scratch resistance, to avoid the occurrence of fissures,
cracks, or canals that invite bacterial growth; (iv) good
adhesion with the hydroxyapatite (HAp) powder par-
ticles and with the substrate (dentin) to avoid micro-
filtration, which produces bacterial growth
29,32,33
;
(v) high hydrolytic stability
3436
for durability within
the environmental conditions of the mouth; and (vi)
appropriate chemistry, to favor nonaggravating molecu-
lar recognition by the immune system. For all composites
of the general type polymer plus ceramic, the problem
of adhesion between differing components is an impor-
tant challenge in product development.
3740
In turn, ad-
hesion depends on surface and interfacial tension val-
ues.
41
Based on these specifications, we selected a poly-
urethane (PU) monocomponent, cross-linkable at room
temperature, as the agglutinating polymer for our HAp-
based hybrids. PUs have a large variety of structures and
thus a variety of properties, from flexible to rigid.
42,43
Our polyisocyanate cross-linker contains a blocking
functionality that limits its own chemical reactivity,
thereby stabilizing the resin-plus-isocyanate mixture. In
the presence of moisture, part of the blocking structure is
released, producing, in this particular case, CO
2
, while
the rest reacts with OH groups that are present in the
material (e.g., on HAp, the hydroxylated PU resin). The
result of the reaction between the HAp and the polymeric
resin is the production of a network in which HAp par-
ticles act as hubs, with polymer molecules chemically
anchored to their surface and joining different HAp par-
ticles through polymer bridges.
44
This network morphol-
ogy of our chemically reacted hybrid improves the me-
chanical properties compared to physical blends of the
same type of material.
Furthermore, the CO
2
produced during the chemical
reaction that occurs when the resin and cross-linker are
mixed with HAp particles acts as a pore-former, creating
a porous material. Both pore size and pore volume frac-
tion in the material depend on (i) the HAp particle size
(and the surface area of HAp particles that is exposed
when the chemical reactions take place) and (ii) the rela-
tive proportion of polyisocyanate (i.e., the OH groups in
the reactive mixture).
Here we describe the synthesis of new porous HAp-
based materials that can be produced in situ at room
temperature and without the presence of other undesir-
able compounds such as pore-formers or other reaction
sideproducts. At the same time, our process is designed
so that we can control the morphology while also achiev-
ing good adhesive, mechanical, and wear properties.
II. EXPERIMENTAL
A. Synthesis
A rigid PU resin was prepared from a commercially
available aliphatic (acrylic) hydroxylated resin (Reich-
hold, Research Triangle Park, NC) combined with a
malonate-blocked polyisocyanate (prepared in the la-
boratory) as the curing agent (CA) in a proportion of 4:1
by volume (to react all cyano groups). Using the same
resin/CA ratio and the same CA, a flexible PU resin was
prepared from a commercial aromatic hydroxylated resin
(Bayer, Leverkusen, Germany) while a semi-rigid
sample was obtained by mixing equal proportions of the
rigid and flexible resins. The reactions occurred at room
temperature, and chemical compositions of the samples
are reported in Table I.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1588
B. Sample preparation
Small disks of 1 cm diameter and 0.5 cm thickness
were produced at room temperature by mixing synthetic
HAp [Ca
10
(PO
4
)
6
(OH)
2
] powder (Sigma Aldrich, Mil-
waukee, WI) ground at a mesh of 100 (particles 150 m
diameter) with the three different PU monocomponent
resins: rigid; semi-rigid; and flexible. Once a homoge-
neous paste was obtained, the mixture was placed in
Teflon (DuPont, Wilmington, DE) molds to produce the
disks. Samples were kept in the molds for 24 h to ensure
full curing of the thermoset resins. For some samples,
alumina microparticles (1.4 m diameter) were added (at
20 and 40 wt% with respect to HAp).
C. Staining
By rubbing recently prepared samples with a spatula
or other flat object, the pores of the surface are closed,
leaving a smooth (poreless) surface. Some of the samples
were rubbed until smooth and then soaked for 2 weeks in
an aqueous solution of methylene blue (MB), a common
staining agent. After that time, the samples were re-
moved from the staining liquid and cut down the middle.
The MB penetration was measured from the cross section
using an optical microscope. The results of the microfil-
tration test are reported in Table I.
D. X-ray diffraction
The x-ray diffraction (XRD) patterns were obtained
using a Rigaku (Tokyo, Japan) D500 machine with a
radiation source of 0.154 nm. The angle 2 was varied
from 10 to 80 at a rate of 2/min.
E. Density
The densities were determined by weighing the
samples on an analytical balance (resolution 1 10
5
g)
and measuring the disk sizes with a micrometer (resolu-
tion 1 10
4
cm).
F. Scanning electron microscopy
Scanning electron microscopy (SEM) was performed
with a JEOL (Tokyo, Japan) JSM-6060 scanning electron
microscope at 20 kV in secondary electron mode with
different magnifications. Samples were frozen in liquid
N
2
and broken, and the new fractured surface was coated
with gold. Average pore sizes were determined from
SEM images by measuring the diameters of >200 pores
for each sample.
G. Tensile mechanical tests
Mechanical tests were performed in an Adamel Lho-
margy (Konkonkoma, NY) machine model DY.22 at
2.5 cm/min. The experiments were performed (five times
for each sample type) in a three-point-bending configu-
ration with two supports separated by 2.0 cm and the
pressure applied in the middle by the probe.
H. Abrasion testing
Abrasion experiments were performed on a specially
designed apparatus using the Taber method according to
the ASTM-D-1242-95 standard, using a steel disk of
25.5 cm radius rotating at 250 revolutions per minute
(rpm) with sandpaper Fandeli (Houston, TX) F-120 ad-
hered on its surface. Samples were placed on the rotating
disk, and a load of 10 g was applied; every 20 s, the
weight loss was determined using an analytical balance
(resolution 1 10
5
g). After the sanding process and
before the weight determination, the sanded surface was
wiped clean with a dry, soft cloth while the sandpaper
was cleaned with a soft brush to remove dust. All ex-
periments were performed at room temperature under dry
conditions for a total time of 200 s. Five sets of abrasion
experiments were performed for each sample.
I. Scratch resistance
Scratch resistance was determined on a microscratch
tester (CSEM Instruments, Neuchatel, Switzerland)
equipped with a Rockwell diamond tip (radius 200 mm,
included angle 120). The procedure has been described
before.
4549
During single scratch tests, the instantaneous
penetration depth R
p
was recorded over the length of
5-mm-long scratch grooves. Scratching was conducted at
5 mm/min under applied loads between 5 N and 25 N.
The residual depth R
h
of the scratch groove was recorded
afterward under an applied load of 0.03 N to measure the
amount of viscoelastic recovery (f). The percentage f was
quantified according the following equation, which was
first defined in Ref. 45 (see also Ref. 46):
f =

1
R
h
R
p

100% . (1)
The reported values of R
p
and R
h
are averages of the
values obtained between 0.5 and 4.5 mm along the
scratch groove.
J. Sliding wear
As with the single scratch tests defined immediately
above, sliding wear tests were conducted using the same
TABLE I. Chemical composition and physical properties of the pre-
pared samples.
Samples
HAp
(g)
PU
resin
(g)
Alumina
(g)
Density
of solid
material
(g/cm
3
)
Density
of porous
material
(g/cm
3
)
Stain
penetration
(mm)
R-100 6.0 5.0 0.0 1.547 0.762 0.46
R-100-201 4.8 5.0 1.2 1.528 0.627 0.00
R-100-40 3.6 5.0 2.4 1.541 0.817 0.00
SR-100 6.5 5.0 0.0 1.545 1.004 0.40
F-100 7.0 5.0 0.0 1.542 1.141 0.22
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1589
microscratch tester, similarly equipped, and at the same
sliding speed of 5 mm/min. The principles of sliding
wear determination and the resulting (in most cases)
strain hardening have been described before.
5052
The
test consists of multiple (usually 15) consecutive
scratches across the same groove during which the R
p
and R
h
are recorded for each pass. Such repetitive passes
of the indenter simulate the wear that occurs in real ap-
plications. As with tests of scratch resistance, the values
reported are averages of values obtained along the 5-mm-
long scratch groove between 0.5 and 4.5 mm. We recall
that the percentage f defined by Eq. (1) that was obtained
from sliding wear tests has been related to brittleness B,
which was defined in Ref. 52. The resulting exponential
relation between B and f has been found valid for all
polymers despite vast differences in their properties.
K. Friction
Friction was evaluated using a Nanovea pin-on-disk
tribometer (Microphotonics, Irvine, CA) equipped with a
silicon nitride pin. Samples were tested under an applied
force of 2 N during 5000 revolutions at 100 rpm.
III. CHEMICAL AND
PHYSICAL CHARACTERIZATION
A. Structural stability
The XRD patterns of HAp-plus-resin and HAp-plus-
resin-plus-alumina hybrids were compared to the pat-
terns for pure HAp and alumina. The patterns show that
the chemical reaction between the HAp and the isocya-
nate does not modify the crystalline structure of HAp or
the alumina. All of the characteristic reflections of HAp
and alumina appear in the hybrid material, indicating
unaltered crystalline structures.
Because our purpose was to use the hybrids as obtu-
ration materials, the hydrolytic stability and morphology
of the samples are important properties to define. The
sample names are preceded by R (rigid), SR (semi-rigid),
and F (flexible), with compositions as defined in Table I.
We have determined that when pore size is >2.9 m and
porosity is around 27% by volume, the whole system is
percolated with an interconnected structure of pores that
potentially can be appropriate for vascularization and
therefore for the regeneration of teeth.
B. Morphology of porous samples
The rigid, semi-rigid, and flexible hybrids all possess
a porous morphology. We see in the R-100 sample in
Fig. 1(a) large interconnected pores in the range from 90
to 350 m (mean size 138 76 m, Fig. 2). We note that
a Serbian group has reported
53
a minimum pore size of
0.4 m. The effects of pore size on mechanical properties
will be discussed in Sec. IV. B. We found similar results
in Fig. 1(c) for the rigid sample containing 20% alumina
(R-100-20). The pore size of R-100-20 ranges from 150
to 370 m, but the mean size is 249 98 m. In addition
to the large interconnected pores, there are many small
cl osed pores (<50 m) t hat may reduce t he
FIG. 1. SEM micrographs of all samples at 20: (a) R-100; (b) R-100-20; (c) R-100-40; (d) F-100; and (e) SR-100.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1590
mechanical properties and are not suitable for vascular-
ization. Sample R-100-40 in Fig. 1(c) appears similar
to the R-100 sample, with large pores between 100 and
300 m and a mean size of 169 65 m. Figure 1(c) also
indicates that the R-100-40 sample contains fewer small
closed pores than the R-100-20 sample.
By contrast, we found small pores in the sample seen
in Fig. 1(d) that were uniformly distributed along the
entire sample with just a few large interconnected pores
for the flexible material F-100. The pores in the F-100
material range in size from 100 to > 400 m with a mean
size of 171 99 m. Thus, the average pore size reflects
the abundance of small pores combined with only a few
larger ones. Not surprisingly, the morphology of the
semi-rigid sample SR-100 reflects its flexible and rigid
components. In Fig. 1(e), large interconnected pores are
clearly visible; their sizes range from 100 to > 300 m,
with a mean size of 175 84 m. Although this numeri-
cal average is similar to that for the R-100-40 sample, we
observe that the SR-100 sample has significantly more
closed (blind) pores that do not contribute to the vascu-
larization.
C. Density and volume fraction of pores
As stated earlier, it is known that the pore volume
fraction must be about 27% to percolate the system. The
volume fraction of pores
p
in the prepared samples was
calculated from the densities of a solid nonporous (
s
)
sample and a porous (
p
) sample of the same composition
according to the relationship:

p
= 1

p

s
. (2)
Nonporous samples were prepared in the same way as
described above for porous samples but using an isocya-
nate without the blocking group, thereby eliminating the
formation of CO
2
. The densities listed in Table I were
calculated from the mass, as measured on an analytical
balance, divided by the volume (determined by measur-
ing the sample size on a micrometer).
The pore volume fractions calculated from Eq. (1) are
also reported in Fig. 2 (along with pore size). We found
that the mean pore volume fraction was smallest (26
0.3%) for the F-100 sample. The remaining samples have
higher
p
values that lie between 35% and 59%, all of
which are suitable in terms of their potential for allowing
vascularization.
D. Stain susceptibility
The microfiltration of the MB stain does not appear to
directly correlate with either pore size or pore volume.
The data in Table I indicate that stain penetration was
immeasurable for the rigid samples containing alumina
and <0.5 mm for the other sample compositions.
IV. MECHANICAL AND
TRIBOLOGICAL CHARACTERIZATION
A. Abrasion
Because the mastication process imposes significant
shear stresses that scratch and abrade the teeth, it is es-
sential to obtain a filling material with sufficient dura-
bility. The results of our abrasion experiments are re-
ported in Fig. 3(a), where weight loss is plotted as a
function of abrasion time (up to 200 s for all samples).
The dependence of weight loss on abrasion time is nearly
linear, with the slopes of the lines corresponding to the
time rate of weight loss [values in Fig. 3(b)]. The semi-
rigid sample SR-100 performed the worst, losing weight
at a rate of 0.063 mg/s (or 3.78 mg/min). With a weight
loss rate of 0.055 mg/s, the R-100 sample performed
somewhat better. The addition of alumina provides an
improvement: Fig. 3(b) shows rates of 0.046 mg/s and
0.047 mg/s, respectively, for the R-100-20 and R-100-40
samples. These two samples have practically the same
abrasion resistance despite a difference in pore size and
pore volume fraction. Apparently, doubling the alumina
concentration has a negligible effect. [The errors for the
rates of weight loss are provided in Fig. 3(b).]
The sample with the highest abrasion resistance was
the F-100 sample, which lost weight at 0.032 mg/s: this
value is practically half that of the semi-rigid sample. In
this scenario, a flexible resin is beneficial because softer
materials are more difficult to grind and generate less
particulate debris. Furthermore, the presence of several
large pores with diameters >300 m may contribute to
the poor abrasion resistance of the SR-100 sample. More-
over, as the subsequent section on mechanical properties
suggests, the rigid component of the SR-100 sample
seems to have more influence on the sample properties
than does the flexible component (at the concentration
used).
FIG. 2. Values of the pore size and pore volume fraction for all
samples.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1591
B. Mechanical properties
We report the outcome of mechanical tests in Fig. 4.
For all samples, the Youngs modulus is fairly low as a
result of porosity in the materials. This is a widespread
effect in porous materials compared to their nonporous
counterparts. The pores create many points for stress
concentration, while their thin walls favor crack propa-
gation and fissures as a result of the high local stresses.
We recall molecular-dynamics computer simulations of
polymeric materials: small cracks can grow into large
ones (unless crack arrest occurs).
54,55
However, some
investigators
56
have demonstrated that the Youngs
modulus can be increased significantly when an organic
phase (e.g., collagen) fills the interconnected pores. Here
also, we see the clear effects of pores and of the com-
position variation. The F-100 sample has the lowest
modulus by far for two reasons: it has the lowest
p
; and
the flexibility of the PU.
As expected based on our previous work,
29,45
tough
ceramic particles improve the mechanics as well as the
tribology of the hybrid materials, as manifested in the
moduli and also in the improved abrasion resistance.
Thus, the two samples with Al
2
O
3
are the strongest. The
R-100-40 sample, which has the most alumina, has the
highest modulus of all; both the R-100-20 and R-100-40
samples contain the same amount of PU. Quite pertinent
also is the result for the R-100 sample; its modulus is
only slightly lower than that of the R-100-20 sample.
This means that the reinforcement effect caused by a
rigid PU filling large pores is much larger than that
caused by the presence of Al
2
O
3
. However, return now to
Fig. 3(b). The alumina-containing samples have lower
abrasion rates than the other samples, except for the
F-100 sample, which has small pores and a flexible PU.
Thus, if optimization with respect to both modulus and
abrasion resistance is needed, the R-100-20 sample
would be the material of choice.
C. Scratch resistance and sliding wear
We see in Fig. 5 the penetration R
p
and residual depth
R
h
values corresponding to single scratch tests at forces
between 5 N and 25 N. For all samples, both R
p
and R
h
increase with increasing load, as expected. While the
responses of some types of materials are linear, the wavy
nature of the R
p
and R
h
curves for other materials is due
to porosity and the particulate nature of our materials.
For instance, the maxima observed for the F-100 and
SR-100 samples can come from hitting a thin-walled
pore, where the diamond indenter is able to penetrate
more deeply. Furthermore, we observe that, overall, the
R
p
is the highest in the flexible material while the R
h
is
simultaneously the most shallow. A softer material is
more easily penetrated but experiences greater viscoelas-
tic recovery f. The rigid sample R-100 has the smallest R
p
and also better f than the sample containing alumina
(sample R-100-20). We recall the results for polyamide-6
FIG. 3. Abrasion results: (a) weight loss versus abrasion time and (b) weight loss rate for all samples. Standard deviation is reported for n 5.
FIG. 4. Youngs modulus for all samples. Standard deviation is re-
ported for n 5.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1592
reinforced with multiwalled carbon nanotubes (CNTs);
CNTs provide mechanical strengthening but hamper the f.
57
The semi-rigid material exhibits intermediate values, al-
though it tends more toward rigid behavior than flexible.
In tests of sliding wear, we evaluate the behavior of the
obturation materials during repetitive scratching. The
depths are plotted as a function of the number of passes
of the indenter: after about 10 passes, the beginning of a
plateau is observed in the R
p
values (Fig. 6); however,
values of R
h
appear to continue to worsen. Generally, the
observance of strain hardening in polymer-based materi-
als (PBMs) is defined by the asymptotic behavior of both
R
p
and R
h
curves.
50,52
For our materials, the behavior
observed here suggests that repetitive wear does not sig-
nificantly increase penetration but damages the material
such that it recovers less with each pass of the indenter.
Overall, the R
p
was highest for the F-100 sample (Figs.
5 and 6), while for sliding wear under a 5 N applied load
the R-100 sample has the most shallow R
p
. Values for the
semi-rigid and rigid samples without alumina are some-
what higher. This is similar to the cluster of values at a
load of 5 N seen above in Fig. 5 for R
p
after single
scratching. Most important is the f of the samples. Simi-
lar to the results for single scratches (at a load of 5 N), in
sliding wear the SR-100 sample has the lowest R
h
, while
the F-100 and R-100 samples have slightly higher values
and the R-100-20 sample has the highest value. Thus, the
highest concentration of alumina we used hampered the
f the most. Note that even the highest recorded value of
R
h
for sample R-100-20 is approximately 140 m, which
is less than or equal to the pore diameters and the HAp
particle sizes in all the samples.
D. Friction
The friction behavior of dental obturation materials is
relevant in determining the durability of such materials.
Figure 7 shows that friction is higher for the flexible and
semi-rigid materials and as low as 0.4 for samples con-
taining the rigid PU. Because we know that friction is
related to other material properties, we are not surprised
to find that the friction values support the results of the
mechanical properties obtained and discussed earlier. We
infer that the higher elastic modulus of the rigid materials
plays a significant role in their lower friction.
V. A SURVEY OF RESULTS
New HAp-based hybrid porous materials were pro-
duced with morphology and mechanical properties ap-
propriate for use as vascularized obturation materials.
FIG. 5. R
p
and R
h
during single scratches plotted versus the force
applied.
FIG. 6. R
p
and R
h
from sliding wear for each pass of the indenter.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1593
This represents a step forward in the design of dental
materials because existing obturation materials are de-
signed to be inert implants, while our materials are de-
signed to actively encourage tooth attachment and regen-
eration. Our new biomaterials have interconnected pores
with average pore size between 100 and 250 m, and
pore volume fractions ranging from 26% to 60%, thus
making them viable candidates to simulate real teeth. The
pore size and volume can be modified by changing the
particle size of the HAp powder and the relative concen-
tration of HAp/resin. This occurs because CO
2
formation
depends on the availability of reactive groups; in turn,
that availability depends on the resin/CA ratio; the
amount of resin needed is dependent on the HAp particle
size. As we establish a quantitative relationship between
these components and the resultant morphology, we will
improve our ability to predict and control the exact mor-
phology of our materials.
While the flexible resin sample F-100 exhibits the best
abrasion resistance and viscoelastic recovery f after
scratching, and also possesses a pore size and volume
fraction near the desired range, its weak mechanical
properties discourage its use as a dental obturation fill-
ing. The semi-rigid sample SR-100 does not match the
mechanical properties of the rigid materials and also ex-
hibits poor resistance to abrasion, thereby negating its
usefulness for the intended application.
The samples with the best performance are R-100,
R-100-20, and R-100-40. Because their pore size is
around 150 m, the R-100 and R-100-40 samples pos-
sess the right morphology; a pore volume fraction near
50% should result in sufficient interconnection for vas-
cular growth. The mechanical properties of the R-100
and R-100-40 samples are combined with a high abrasion
resistance (5.5 and 4.7 10
5
g/s, respectively). Low
friction along with good resistance to scratching, and
decent f after single scratches and sliding wear suggest
the usefulness of the rigid hybrids for obturation materi-
als in dentistry. Given the wide range of rigidity available
in PUs and our capability to vary concentrations of HAp
and alumina, fine tuning of the desired properties is pos-
sible.
Finally, we note the issue of the sources of HAp. In
this work, we have used a synthetic HAp. We recall that
HAp can also be made from egg shells,
58
which other-
wise become an environmental problem.
ACKNOWLEDGMENTS
The authors are in debt to Mrs. Alicia del Real for her
valuable help in SEM analysis and to Mrs. Maribel Presa
for the determination of the mechanical properties. Fi-
nancial support to H.E.H. Lobland has been provided by
a National Defense Science and Engineering Graduate
(NDSEG) Fellowship (Washington, DC).
REFERENCES
1. D. Stojanovic, B. Jokic, Dj. Veljovic, R. Petrovic, P.S. Uskokovic,
and Dj. Janackovic: Bioactive glass-apatite composite coating for
titanium implant synthesized by electrophoretic deposition.
J. Eur. Ceram. Soc. 27, 1595 (2007).
2. C. Balzsi, F. Wber, Z. Kvr, E. Horvth, and C. Nmeth:
Preparation of calcium-phosphate bioceramics from natural re-
sources. J. Eur. Ceram. Soc. 27, 1601 (2007).
3. J.M. Anderson and J.J. Langone: Issues and perspectives on the
biocompatibility and immunotoxicity evaluation of implanted
controlled release systems. J. Control. Release 57, 107 (1999).
4. M. Tanahashi, T. Yao, T. Kokubo, M. Minoda, T. Miyamoto,
T. Nakamura, and T. Yamamuro: Apatite coated on organic poly-
mers by biomimetic process improvement in its adhesion to sub-
strate by NaOH treatment. J. Appl. Biomater. 5, 339 (1994).
5. P.N. de Aza, F. Guitin, and S. de Aza: A new bioactive material
which transforms in situ into hydroxyapatite. Acta Mater. 46,
2541 (1998).
6. Y. Shikinami and M. Okuno: Bioresorbable devices made of
forged composites of hydroxyapatite (HA) particles and poly-L-
lactide (PLLA): Part I. Basic characteristics. Biomaterials 20, 859
(1999).
7. M.E. Thomas, P.W. Richter, T. van Deventer, J. Crooks, and
U. Ripamonti: Macroporous synthetic hydroxyapatite bioceramics
for bone substitute applications. S. Afr. J. Sci. 95, 359 (1999).
8. R. Schilke, J.A. Lisson, O. Baub, and W. Geurtsen: Comparison of
the number and diameter of dentinal tubules in human and bovine
dentine by scanning electron microscopic investigation. Arch.
Oral Biol. 45(5), 355 (2000).
9. R.S. Manly, H.C. Hodge, and L.E. Ange: Density and refractive
index studies of dental hard tissues: II. Density distribution curves.
J. Dent. Res. 15, 203 (1939).
10. S.A. Patterson: In vivo and in vitro studies of the effect of diso-
dium salt of ethylenediamine tetra-acetate on human dentine and
its endodontic implications. Oral Surg. 16(1), 83 (1987).
11. D.H. Pashley: Dentin-predentin complex and its permeability:
Physiology overview. J. Dent. Res. 64, 613 (1985).
12. L.B. Goldman, M. Goldman, J.H. Kronman, and P.S. Lin: The
efficacy of several irrigations solutions for endodontics: A scan-
ning electron microscopy study. Oral Surg. 52(2), 197 (1985).
FIG. 7. Friction of the samples as determined by a pin-on-disk tri-
bometer.
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1594
13. A. Cosijns, C. Vervaet, J. Luyten, S. Mullens, F. Siepmann,
L. Van Hoorebeke, B. Masschaele, V. Cnudde, and J.P. Remon:
Porous hydroxyapatite tablets as carriers for low-dosed drugs.
Eur. J. Pharmacol. Biopharmacol. 67(2), 498 (2007).
14. D. Gupta, S. Chandra, and S. Chandra: Effect of the smeared layer
upon dentinal tubule penetration by root canal sealers: A SEM
study. Endodontology 8(1), 26 (1996).
15. S. Teng, L. Chen, Y. Guo, and J. Shi: Formation of nano-
hydroxyapatite in gelatin droplets and the resulting porous com-
posite microspheres. J. Inorg. Biochem. 101, 686 (2007).
16. J.F. Piecuch: Long-term evaluation of the use of coralline hy-
droxyapatite in orthognathic surgery. J. Oral Maxillofac. Surg. 56,
941 (1998).
17. N. Ozgr Engin and A. Cney Tas: Manufacture of macroporous
calcium hydroxyapatite bioceramics. J. Eur. Ceram. Soc. 19, 2569
(1999).
18. L.K.A. Rodriguez, M.N. dos Santos, D. Pereira, A.V. Assaf, and
V. Pardi: Carbon dioxide laser in dental caries prevention. J. Dent.
32, 531 (2004).
19. H.J. Lee, S.E. Kim, H.W. Choi, C.W. Kim, K.J. Kim, and
S.C. Lee: The effect of surface-modified nano-hydroxyapatite on
biocompatibility of poly (E-caprolactone)/hydroxyapatite
nanocomposites. Eur. Polym. J. 43, 1602 (2007).
20. P.S. Uskokovic, C.Y. Tang, C.P. Tsui, N. Ignjatovic, and
D.P. Uskokovic: Micromechanical properties of a hydroxyapatite/
poly-L-lactide biocomposite using nanoindentation and modulus
mapping. J. Eur. Ceram. Soc. 27, 1559 (2007).
21. J.P. Paul: Strength requirements for internal and external prosthe-
ses. J. Biomech. 32, 381 (1999).
22. M. Spector: Historical review of porous-coated implants. J. Ar-
throplasty 2, 163 (1987).
23. P.H.F. Caria, E.Y. Kawachi, C.A. Bertran, and J.A. Camilli: Bio-
logical assessment of porous-implant hydroxyapatite combines
with periosteal grafting in maxillary defects. J. Oral Maxillofac.
Surg. 65, 847 (2007).
24. M. Rabello: Putting Additives into Polymers (Artliber, Sao Paulo
2000), Chap. 10.
25. L.H. He and M.V. Swain: Enamel: A metallic-like deformable
biocomposite. J. Dent. 35, 431 (2007).
26. C-L. Lin, S-H. Chang, J-C. Wang, and W-J. Chang: Mechanical
interactions of an implant/tooth-supported system under different
periodontal supports and number of splinted teeth with rigid and
non-rigid connections. J. Dent. 34, 682 (2006).
27. B. Fischman: The rotational aspect of mandibular flexure.
J. Prosth. Dent. 64, 483 (1990).
28. A. de la Isla, W. Brostow, B. Bujard, M. Estevez, J.R. Rodriguez,
S. Vargas, and V.M. Castao: Nanohybrid scratch resistant coat-
ings for teeth and bone viscoelasticity manifested in tribology.
Mater. Res. Innovat. 7(2), 110 (2003).
29. M. Estevez, S. Vargas, H.E. Hagg Lobland, A. de la Isla,
W. Brostow, and J. Rogelio Rodrguez: Characterization of novel
dental obturation materials. Mater. Res. Innovat. 10(4), 411
(2006).
30. V.M. Castao and R. Rodriguez: Polymer-based hybrid organic-
inorganic materials, in Performance of Plastics, edited by
W. Brostow (Hanser, Munich, 2000), Chap. 24.
31. M. Deng and S.W. Shalaby: Polymers as biomaterials, in Perfor-
mance of Plastics, edited by W. Brostow (Hanser, Munich, 2000),
Chap. 23.
32. C. Hengtrakool, G.J. Pearson, and M. Wilson: Interaction between
GIC and S Sanguis biofilms: Antibacterial properties and changes
of surfaces hardness. J. Dent. 34, 588 (2006).
33. T.J. Griffin and W.S. Cheung: The use of short, wide implants in
posterior areas with reduced bone height: A retrospective inves-
tigation. J. Prosth. Dent. 92, 139 (2004).
34. J.W. Freudenthaler, G.K. Tischler, and C.J. Burstone: Bond
strength of fiber-reinforced composite bars for orthodontic attach-
ment. Am. J. Orthod. Dentofacial Orthop. 120, 648 (2001).
35. F. Tancret, J-M. Bouler, J. Chamousset, and L-M. Minois: Mod-
elling the mechanical properties of microporous and macroporous
biphasic calcium phosphate bioceramics. J. Eur. Ceram. Soc. 26,
3647 (2006).
36. C.E. Aragon and R.N. Bohay: The application of alveolar distrac-
tion osteogenesis following nonresorbable hydroxyapatite grafting
in the anterior maxilla: A clinical report. J. Prosth. Dent. 93, 518
(2005).
37. R.P. Wool: Interfaces and adhesion, in Performance of Plastics,
edited by W. Brostow (Hanser, Munich, 2000), Chap. 15.
38. R. Mlhaupt: Toughened thermoplastics and thermosets, in Per-
formance of Plastics, edited by W. Brostow (Hanser, Munich,
2000), Chap. 20.
39. Z. Roslaniec, G. Broza, and K. Schulte: Nanocomposites based on
multiblock polyester elastomers (PEEs) and carbon nanotubes
(CNTs). Compos. Interfaces 10, 95 (2003).
40. A. Bismarck, M. Hofmeier, and G. Drner: Effect of hot water
immersion on the performance of carbon reinforced unidirectional
poly(ether ether ketone) (PEEK) composites: Stress rupture under
end-loaded bending. Composites A 38, 407 (2007).
41. A. Kopczynska and G.W. Ehrenstein: Polymeric surfaces and
their true surface tension in solids and melts. J. Mater. Ed. 29, 325
(2007).
42. W. Brostow, J.V. Duffy, G.F. Lee, and K. Madejczyk: Parameters
of equation of state of polyurethanes from acoustic resonance and
isobaric expansivity. Macromolecules 24, 479 (1991).
43. E.B. Mano: Polymers as Engineering Materials (Edgard Blcher,
So Paulo, 1996).
44. L.G. Griffith: Polymeric biomaterials. Acta Mater. 48, 263 (2000).
45. W. Brostow, B. Bujard, P.E. Cassidy, H.E. Hagg, and
P.E. Montemartini: Effects of fluoropolymer addition to an epoxy
on scratch depth and recovery. Mater. Res. lnnovat. 6, 7 (2001).
46. W. Brostow, J-L. Deborde, M. Jaklewicz, and P. Olszynski:
Tribology with emphasis on polymers: Friction, scratch resistance
and wear. J. Mater. Ed. 25, 119 (2003).
47. W. Brostow and M. Jaklewicz: Tribology of a polymeric molecu-
lar composite: Effects of magnetic field orientation. J. Mater. Res.
19, 1038 (2004).
48. M. Estevez, S. Vargas, V.M. Castao, J.R. Rodriguez,
H.E. Hagg Lobland, and W. Brostow: Novel wear resistant and
low toxicity dental obturation materials. Mater. Lett. 61, 3025
(2007).
49. W. Brostow, W. Chonkaew, L. Rapoport, Y. Soifer, and
A. Verdyan: Grooves in microscratch testing. J. Mater. Res. 22,
2483 (2007).
50. W. Brostow, G. Damarla, J. Howe, and D. Pietkiewicz: Determi-
nation of wear of surfaces by scratch testing. e-Polymers 025, 1
(2004).
51. M.D. Bermudez, W. Brostow, F.J. Carrion-Vilches, J.J. Cervantes,
and D. Pietkiewicz: Wear of thermoplastics determined by mul-
tiple scratching. e-Polymers 005, 1 (2005).
52. W. Brostow, H.E. Hagg Lobland, and M. Narkis: Sliding wear,
viscoelasticicity, and brittleness of polymers. J. Mater. Res. 21,
2422 (2006).
53. L. Vujosevic and K. Obradovic-Durici: Porosity of hard dental
tissues. Stomatol. Glasnik Serb. 36(2), 95 (1989).
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1595
54. W. Brostow, A.M. Cunha, J. Quintanilla, and R. Simes: Predict-
ing cracking phenomena in molecular dynamics simulations
of polymer liquid crystals. Macromol. Theory Simul. 11, 308
(2002).
55. W. Brostow and R. Simes: Tribological and mechanical behavior
of polymers simulated by molecular dynamics. J. Mater. Ed. 27,
19 (2005).
56. E. Rivera-Muoz, R. Velzquez, and R. Rodrguez: Improvement
in mechanical properties of hydroxyapatite objects with controlled
porosity made by modified gelcasting process. Mater. Sci. Forum
426432, 4489 (2003); on line at http://www.scientific.net.
57. L.F. Giraldo, W. Brostow, E. Devaux, B.L. Lopez, and L.D. Perez:
Scratch and wear resistance of polyamide 6 reinforced with mul-
tiwall carbon nanotubes. J. Nanosci. Nanotechnol. 8, (2008, in
press).
58. E.M. Rivera, M. Araiza, W. Brostow, V.M. Castao,
J.R. Diaz-Estrada, R. Hernndez, and J.R. Rodrguez: Synthesis of
hydroxyapatite from eggshells. Mater. Lett. 41, 128 (1999).
W. Brostow et al.: Porous hydroxyapatite-based obturation materials for dentistry
J. Mater. Res., Vol. 23, No. 6, Jun 2008 1596